University of Southampton Intramolecular Radical Additions to Aromatic Compounds By Nigel James Blumire A Thesis submitted for the degree of Doctor of Philosophy Department of Chemistry Faculty of Science University of Southampton ABSTRACT FACULTY OF SCIENCE CHEMISTRY Doctor of Philosophy Intramolecular radical additions to aromatic compounds By Nigel James Blumire This thesis is concerned with the development of cyclisation strategies that allow entry to polycyclic heterocycles and medium sized ring systems. Radical additions to pyridines are developed and methodologies explore the differences between the homolysis of C-Br and C-I bonds. Mechanisms are described for the unexpected ipso additions to the pyridines. Work towards the synthesis of the stegnans and dimethylgomisin illustrates the attempts to utilise the ipso addition to form medium sized ring systems. A new method of synthesising these medium sized ring systems is discussed. Progress towards a general methodology is described and shows the opportunities for which it could be utilised. A literature review of the synthesis of heterocycles via radical cyclisations is presented. Contents Chapter One Introduction The Synthesis of heterocycles by radical cyclisations 1.0 Introduction 2 1.1. Synthesis of Heterocycles by Radical Cyclisation 2 1.1.1. Synthesis of Pyrrolidines and Piperidines 2 1.1.2. Synthesis of Related Condensed Heterocycles 5 1.1.3. Natural Product Synthesis 10 1.2. Medium Sized Heterocycles 15 1.2.1. Introduction 15 1.2.2. Alkoxy Radical Fragmentations 15 1.2.3. Aromatic Radical cyclisations 16 1.2.4. Tandem Radical Cyclisations 17 1.3. Radical Additions to Nitrogen Containing Heterocycles 18 1.4. Conclusions 21 Chapter Two Radical Additions to Pyridines 2.1. Background 23 2.2. Aims of Investigation 25 2.3. Additions to Pyridines 26 2.3.1. Bromides 26 2.3.2. Iodides 32 2.4.Conclusions 37 Chapter Three The Formation of Medium Ring Systems: an Intramolecular Radical Approach 3.1. Background 39 3.2. Formation of Medium Ring Systems 43 3.3. Aims of Investigation 46 3.4. Stegnans 47 3.5. Dimethylgomisin 49 3.6. Intramolecular Radical Approach to Medium Sized Rings 51 3.7. Tetralins 58 3.8. Alternative Radical Generating Methods 61 3.9. Conclusions 62 3.8, Further Work 62 Chapter Four Experimental 4.1, General Experimental 64 4.2. Experimental 66 Chapter Five References 163 Appendix Appendix A: l,2-Dimethoxy-7-oxo-5,6,7,8-tetrahydrodibenzo[a,c]cyclooctene-6- carboxylic acid methyl ester 3.98 Appendix B; X-ray data for Z-3-[(2-Bromo-l,3-benzodioxol-5-yl)-l-ethenyl]pyridine 2.28 Appendix C: 3,4-Dimethoxy-6-methyl-2'-(2-hydroxy malonic acid monomethyl ester)-biphenyl 3.125 Preface The research described in this thesis was carried out under the supervision of Dr. D.C. Harrowven at the University of Southampton between October 1999 and October 2002. No part of this thesis has previously been submitted for a degree. Acknowledgements Firstly I would like to thank David Harrowven and the University of Southampton for the opportunity to do this research and for having faith in my abilities when I had so many doubts myself I would like to thank David and John Mellor for all their advice and help during the past three years. Four people that I need to thank for all their hard work are Neil Wells, Joan Street, John Langley and Julie Hemiman for everything they do. I can't thank them enough for the time they have spent keeping the spectrometers running and helping me. Thanks also need to be said to my family for all their support during my extended time at university. Sorry it took so long but we are almost there now and hopefully I can back all of the time and patience expended on me. Finally I wish to thank some very important people. The group I have worked in and the friends around me. I especially wish to mention Michael Nunn, Mathieu Carpentier, Ben Sutton, Heather Bourne, Stuart Flanagan, Melloney Tyte, Tim Woodcock. Thanks to them for all of the ^reat times we have spent together over the years. Thank you also to all of my friends especially the guys down the Gate. I hope that they're still winning. There are some other people that I should thank for their encouragement and help. Firstly I would like to thank Maugan Higgins for his help and advice over the years. I would also like to thank two mentors of mine, Dr Spencer and Farid Azizian. Both of them have helped me to fiilfil a dream that they helped to form in my mind. Thank you for being there even when I wasn't. "It is a strange fate that we should suffer so much fear and doubt over so small a thing. Such a little thing." Boromir, The Fellowship of the Ring Abbreviations AIBN azo-/w-butyronitrile amu atomic mass units APCI atmospheric pressure chemical ionization aq. aqueous AJ aryj Bn benzyl Bu butyl CHN combustion analysis CI chemical ionisation conc. concentration COSY correlated spectroscopy DBU 1,8-diazabicyclo[5.4.0]undec-7-ene DCM dichloToraethane DMF TV^TV-dimethylfbrmamide DMSO dimethylsulfoxide EI electron impact eq. equivalents Et ethyl FT Fourier Transform h hours HMPA hexamethylphosphoramide HRMS high resolution mass spectra IR infrared lit, literature LRMS low resolution mass spectra LDA lithium diisopropylamide LiHMDS lithium hexamethyldisilazide M molar mmol jnillimoles Me methyl min minutes NMR nuclear magnetic resonance Ph phenyl ppm parts per million py pyridine THF tetrahydrofuran TBDMS ter^-butyldimethylsilyl TMEDA tetramethylethylenediamine TMS trimethylsilyl UV ultra violet Chapter 1 Introduction 1.0 Introduction Radical cyclisation strategies for the synthesis of heterocycles are now a well established and commonly used. The vast majority of radical cyclisations use tributyltin hydride as a mediator but other methods are becoming more common. This review focuses on the use of radical cyclisations to form nitrogen-containing heterocycles. 1.1 Synthesis of heterocycles by radical cyclisation 1.1.1 Synthesis of pyrrolidines and piperidines The synthesis of pyrrolidines by 5-exo-tng radical cyclisation, provides a common route to these ring systems. The radical can be formed at various positions in relation to the A'-heteroatom. Cyclisation of radicals a- to the 7V-heteroatom have been widely studied and a range of pyrrolidines have been prepared in this way. For example by samarium diiodide mediated 5-exo-trig cyclisation 7V-(a- benzotriazolylalkyl)alkenylamines^ generally proceed smoothly (Scheme 1). N R 1.2 1.3 Reagents and Conditions; a. RCHO, BtH; b. Sm^,, THF, HIVlPA; c. E"^ Bt = 1 - or 2- Benzotriazolyl Scheme 1: Cyclisation of (benzotriazolylalkyl)alkenylamines More popular are radical reactions from the carbon centre P to the nitrogen atom. For instance, a useful synthetic procedure uses Michael type addition reactions of allyl oj propargyl-amines to a-phenylselenide-a,P-unsaturated esters, amides, ketones, nitriles and sulfones to yield radical precursors. The phenylselenide group may then be abstracted by tris(trimethylsilyl)silane with triethylborane as initiator. The resulting radical intermediates then undergo 5-exo-trig (or 5-exo-dig) cyclisations to give the corresponding pyrrolidine (or dihydropyrrole) derivatives (Scheme 2).^ PhSe^ ^EWG EWG^ ,SePh HoN' R N H 1.5 1.6 1.7 a, b EWG EWQ. y R' N Ts Ts 1.9 1.8 Reagents and Conditions: a. (TMS)3SiH, EtgB, Og, PhMe; b. TsCI, EtsN, two diastereoisonriers, 36 - 89 % Scheme 2; Radical cyclisation using phenylselenide groups An example of endo cyclisation leading to piperidines has been reported involving the cyclisation of alkenyl radicals onto an imine/ The alkenyl radical was generated by addition of BusSn' to a terminal alkyne. The resulting alkenyl radical then underwent a 6-endo cyclisation leading to 1.13. The alternative pathway, 5-exo-trig cyclisation to 1.12 and rearrangement to 1.13 was ruled out as cyclisation to the imine nitrogen is disfavoured (Scheme 3), BugSn" 1.10 1.11 X BugSn Bu3Sn"^%^^NH ^ BusSn 1.14 1.13 1.12 Reagents and Conditions: a. BusSnH, AIBN, PhH, A, 27 - 43 % Scheme 3: Radical cyclisation using the addition of tributyltin radical Clark et al. have shown that copper(I) reagents can may be used instead of toxic tributyltin hydride in some instances. Formation of di- and tetrahydroindanones by 5- 3 endo-tn% radical cyclisation provides an excellent illustration of this powerful methodology (Scheme 4)/ Me_ Me. ,Br O O^H Bn 1.15 1.16 1.17 CI CI CI CI o O^N N Bn Bn 1.18 1.19 Reagents and Conditions; a. 30 mol% CuBr{ligand 1), DCM, rt, 20 mins; b. 1 eq. CuCI {ligand 2), DCM, 40°C, 48 hrs Me2N. NMe2 Ligand 1= N NMe? Ligand 2= Scheme 4; Use of copper mediated reactions Cyclisation involving iV-centered radicals have also been investigated by Clark et al. In most cases only a small amount of reduced material is isolated.^ Yields are moderate in most cases but the stereoselectivity of these reactions show up effectiveness of radical cyclisation reactions. R R O R ,R' N-R' + ^^I^N-R' V" O O o 1.20 1.21 1.22 1.23 Reagents and Conditions; a. BugSnH, AIBN, PhMe, 110°C, 8 hrs Scheme 5: 5-exo-trig cyclisations involving vV-centered radicals R R' Cyclised/reduced Yield of 1.22 +1.23 d.e. Me Me 17:1 47% 17% Me Bn 12:1 53% 23% Ph Me 3.5:1 22% 43% Table 1: Results of radical cyclisation 1.1.2 Synthesis of related condensed heterocycles The cyclisation of aryl radicals onto unsaturated side chains containing nitrogen provides one of the most commonly used radical methodologies for the synthesis of nitrogen heterocycles.